IoT Platform using Microservices - SPRC

Name: Dumitrescu Alexandra
Date: November 2023

Content

1. Project Structure
2. Implementation
   1. Microservices
   2. Run & cleanup commands
   3. Client tester & Testing
3. Links

Project Structure

server/ - Implements the adaptor used for parsing messages received from the MQTT broker and adding them
to the Influx-DB database
influx-db/ - Initial script for creating the database where the measurements will be stored IoTSPRC
client/UPB - Testing script
grafana/ - dashboards/ - JSON used for creating the Battery Dashboard and UPB IoT Dashboard
provisioning/ - .yml files used for connecting to the Influx DB and setting path to the JSON files for the dashboards

Implementation

Microservices

MQTT Broker on port 1883, used eclipse mosquitto image
Grafana on port 80
InfluxDB in the ./influx-db folder there is an initial script for creating the IoTSPRC database. Adaptor communicates with the MQTT broker and with Influx DB service. Receives message on the following format:

<location/station> topic
single-level JSON payload

All the values in the payload having type int or float will be added in the database.
We propose a solution with common measurement for all data being stored - IoT, where we add the data having the tags: location, station, key, <station.key>
and the value the received data.
The adaptor also prints logging info if DEBUG_DATA_FLOW is set to "true". Run the following command to see the logs:

docker service logs sprc3_adaptor-server

Run & cleanup commands

In the main directory there are 2 bash scripts used for running and cleanup. Before running make sure to set the SPRC_DVP environment variable.
When running the cleanup script, the volumes will be deleted.

Client tester & Testing

In the client/UPB there is a python script used for testing. There are 2 locations and 3 stations:

location = ["UPB", "Test"]
stations = ["Gas", "Mongo", "Tel"]

For each pair of <location/station> I randomly generate float/integer values and strings.
I also send items to a "Dummy" topic in order to check the restriction on the payload.
For testing and simulationg just run the client.py script.

Links

https://grafana.com/docs/grafana/latest/setup-grafana/installation/debian/
https://medium.com/javarevisited/monitoring-setup-with-docker-compose-part-2-grafana-2cd2d9ff017b
https://stackoverflow.com/questions/61097164/docker-stack-deploy-error-response-from-daemon-rpc-error-code-invalidargume
https://www.influxdata.com/blog/getting-started-python-influxdb/
https://www.influxdata.com/blog/how-to-setup-influxdb-telegraf-and-grafana-on-docker-part-1/
https://docs.python.org/3/howto/logging.html
https://stackoverflow.com/questions/2257441/random-string-generation-with-upper-case-letters-and-digits
https://stackoverflow.com/questions/63518460/grafana-import-dashboard-as-part-of-docker-compose
https://community.grafana.com/t/where-is-the-alias-for-flux-queries-please/44348/3
https://stackoverflow.com/questions/54813704/how-to-add-dashboard-configuration-json-file-in-grafana-image